Acoustic Identification of Wild Gray Wolves, Canis lupus, Using Low Quality Recordings

Abstract

Invasive trapping and radio-collaring techniques are currently used by conservation biologists to study the population dynamics of gray wolves (Canis lupus). Previous research has found wolf howls can be used to determine individual identity on high quality recordings from captive animals, offering an opportunity for non-invasive monitoring of packs. We recorded wild wolves in Central Wisconsin to determine the effectiveness of these features in determining individuality in low quality recordings. The wolf howls analyzed were from two adult individuals from separate packs. Using a principle component analysis, maximum frequency and end frequency of the calls were determined to be most individualistic. Using these features in a discriminant function analysis, howls were able to be identified from individuals with 100% accuracy. Gray wolves play an important role in ecosystem maintenance, however, the current monitoring techniques are costly and invasive. The creation of an easily accessible, noninvasive technique that can be used by individuals with a variety of technical backgrounds is necessary to address concerns faced by conservation efforts. To address these issues, all analyses performed used free or low-cost software, making this method of individual identification a useful alternative for conservation biologists.